One of the topics you filed this question under is "Biostatistics". In biostatistics, logistic regression is often used when the outcome variable is dichotomous. You also list "SPSS" as a topic. A Google search on will no doubt yield many hits, including the UCLA "textbook examples".
For logistic regression, bear in mind that you should have 15-20 events per model parameter if you wish to avoid over-fitting. (See the Babyak article linked below for more info on over-fitting.)
One of the topics you filed this question under is "Biostatistics". In biostatistics, logistic regression is often used when the outcome variable is dichotomous. You also list "SPSS" as a topic. A Google search on will no doubt yield many hits, including the UCLA "textbook examples".
For logistic regression, bear in mind that you should have 15-20 events per model parameter if you wish to avoid over-fitting. (See the Babyak article linked below for more info on over-fitting.)
you pick regression model based on the distribution of the dependent variable only not caring predictors. In this case it's GLM with binomial family/ logistic regression.
If the dependent variable is binary, you should performe a logistic regression. The assumption is absence of collinearity between the indipendent ones. You should test if the indipendent variables have low correlation each other.
the alternative choise is the use of segmentation algorithms which have no assumption on the distribution of the variables. You can use these both with continue indipendent variables both with categorial or binary.
Use logistic regression for your analysis. You can also compare the results of the logistic regression with probit model to see which one provides the best results.
Independent variables - one is categorical, the other is dichotomous and dependent variable is also dichotomous.
Your dependent variable should be measured on a dichotomous scale.
Example: The dichotomous variables include gender ie, two groups: "males" and "females", presence of heart disease like., two groups: "yes" and "no". personality type (two groups: "introversion" or "extroversion"), body composition (two groups: "obese" or "not obese"), and so forth.
1. However, if your dependent variable was not measured on a dichotomous scale, but a continuous scale instead, you will carry out multiple regression.
2. If your dependent variable was measured on an ordinal scale, ordinal regression would be a more appropriate starting point.